Frédéric H. (E/ACC)

@eYouSocial Do you plan to create a desktop app version of it?

46,000 on the eYou waitlist. We're closing it at 50,000. 4,000 spots left. After that, the door shuts until launch. The Early Believer badge: exclusive, permanent, never available again after 50k. Launch: May. 👇 Who are you sending your link to? eYou.social

@mercury__agent Tried running Mercury locally: Ollama v2 models aren’t supported, the SDK fallback breaks, provider selection is inconsistent, and the setup process fails to complete reliably. Hard to evaluate the tool in these conditions.

Mercury has been trending at the top all week in total stars, and it’s only been a week since our official release. Thank you all for the support. Our community is growing fast, and this is just the beginning. We’re building agentic AI that feels more personal, more soulful, token-efficient, and fully in your control.

@RoundtableSpace @grok Quel programme est-ce ?

AI CAN NOW TURN IMAGES INTO SVGs WITH WAY LESS FRICTION THAN PEOPLE EXPECTED.

@cocktailpeanut Ok :) Thank you very much.

@_Aeden_ That’s what I’m testing right now :) will share everything

Introducing Pinokio 7 P7 is here, and it is all about AI agents. 1. Interpreter: Let AI agents control your apps, with zero config 2. Assistant: Talk to your apps in chat 3. Memory: AI agents remember your chats around each app, so you can pick up where you left off

@cocktailpeanut For local agent, according your experience, what will you recommend to properly / comfortably run your AI agents locally (hardware setup [GPU/VRAM, RAM, etc.]) ?

> Pinokio AI agents only runs apps inside it, those who are installed by Pinockio This is why now anyone can generate a launcher on the fly using AI directly in pinokio. This is no longer a limitation (except for the cost, which is a fair criticism and which i agree with) > For the rest, you have to use API keys - paying [costly], or free (token limitation per min / day) Local agent support coming soon

@cocktailpeanut Pinokio AI agents only runs apps inside it, those who are installed by Pinockio. For the rest, you have to use API keys - paying [costly], or free (token limitation per min / day) and request loops are consumming. If you don't have a strong hardware, you can't run AI locally.

@_Aeden_ what do you mean?

@EpicClipVault Quel cirque...

Woman goes viral after revealing how she trained for Kung Fu lessons in China.

@DaiAoki30 There is an issue with the less command. The exercise is not valided.

Webブラウザ上で動くTerminal「WebTerm」をβリリースしました🚀 会員登録不要・無料・データはephemeral。壊しても良い完全なsandbox環境のTerminalです。 webterm.app 最近、周りでもAIコーディング前提の開発が当たり前になっていて、コードを書かなくなった、チーム開発のやり方が変わった（人がレビューしなくなった）、という会話ばかりです。 そしてこの流れは少しずつ、エンジニア以外にも波及していると感じます。 「GitHubってどう使うの？」「Claude Codeって非エンジニアでも扱える？」と聞かれる機会が明らかに増えました。実際、Claude Codeを使いこなせる非エンジニアがいたらめちゃくちゃ強いと思う。ひと昔前でいうSQLが書けるマーケター的な。（それよりチョットハードル高そうだけど） 組織全体への導入事例も去年から着実に増えています。 ただし、AI時代にCLIは避けて通れないのに、いきなり自分の端末でTerminalを開いて試すのはこわい。 誤ってデータが消える不安もあるし、その心理的障壁は大きい。 そこでWebTermでは ・データは全てephemeral（保存なし. 一時的なもの） ・安全に壊せる ・会員登録不要 ・無料 環境を作りました。とにかく使い始めるハードルを下げることにこだわりました。 興味はあるけれど何から始めればよいかわからない初心者から、触ったことはあるけれど使いこなせていない人まで幅広く使えるよう設計しています。 まだβ版なので足りない機能やバグもあります。70個くらいのコマンドに対応しているのですが、（初中級者の人をメインに使ってもらいたくて作ったこともあって）必須ではないコマンドのデバッグは追いついていないです。 気軽にバグ報告やフィードバックをもらえると助かります。 一応、標準出力のリダイレクトには対応してるのですが、パイプを利用したコマンドのチェーンとかはちょっと対応が難しくてpendingしてます。（そういう応用的なことをしようとする人はそもそも本物のTerminal使ってください笑） とはいえ、気分転換で今後、そういうところも作り込んでいくかもしれません。 より多くの人が気軽にAI時代のCLIを体験できるようになれば嬉しいです。 他にも作りたいものいろいろあるので、2026年は初心を忘れずコツコツやります。

@ErenChenAI AI videos become very borying

Introducing our newest humanoid model. Built to take on three G1s at once 😂 Just kidding.

@RuairiRobinson @grok File moi le prompt de cette vidéo

This was a 2 line prompt in seedance 2. If the hollywood is cooked guys are right maybe the hollywood is cooked guys are cooked too idk.

@drinkonsaturday " Modern " women want to prove all the time...

She told him, "if you tap out, it's a 'no.'" Looks like my guy locked-in and got himself a date.

@BrianRoemmele @grok Comment a été créée cette vidéo ? Avec quel outil ?

I’m sorry. Doing my best to prepare you for what I have seen ahead.

@SoveyX @Grok Is this real?

I am not paying to stand in the sky like I am waiting for a table at Olive Garden. <I do like Olive Garden tho>

@joshwoodward @GeminiApp When will it be available in France?

We just shipped our biggest update to Gemini Live ever. It's smarter. It's more expressive. It has accents. And my favorite: it can talk faster! Here are 3 ways people are already using it in @GeminiApp. 🧵

@FirstDoctor Gorrila

Simple eye test, what animal can you see?

@itsPaulAi But it's not a vector image.

Dall-E 3 with the new Windows Copilot is amazing. You can generate a vector illustration in 2 clicks and import it into PowerPoint (or Word). Open Windows Copilot and write this prompt to generate an image: E.g.: "Create an image: a chief cooking a raspberry pie, flat simple vector illustrations style, vibrant colors, white background" Then click on the image you prefer to display it in full quality. Just copy and paste it into your presentation. It only takes a few seconds and you don't even have to search the Internet. Follow me for more AI content!

@The_Truepath Forest Gump

Keanu Reeves once said: "I’m at the stage in life where I stay out of arguments. Even if you say 1+1=5, you’re right. Have fun." Here are 9 things I learned from him:

@Noxywoxy What a shame.

@itsPaulAi @JN07435970 It's not the beta, but the dev version. And, for European people, like french, we have to use a VPN (Urban VPN is completely Free).

@JN07435970 Thanks for the confirmation. So it should arrive soon in a stable version. You can install the Beta version of Edge in the meantime.

Copilot Vision can now see anything on your screen. This gives you the power to understand literally everything you see. (I show you how to enable the feature below)